Model comparison
GPT-5.6 Terra vs WizardLM-2 8x22B
GPT-5.6 Terra has enough public results to be ranked (#17); WizardLM-2 8x22B does not yet, so treat this comparison as directional.
Last verified . 1 shared benchmarks.
MicrosoftSummary
- They share 1 benchmark with published results for both. GPT-5.6 Terra scores higher in 2 categories and WizardLM-2 8x22B in 0 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Terra leads 81.6 to 28.9.
- The biggest single-benchmark swing is GPQA Diamond: 93.3% for GPT-5.6 Terra and 43.4% for WizardLM-2 8x22B.
- WizardLM-2 8x22B is cheaper at $0.62 / $0.62 per million input/output tokens, against $2 / $12 for GPT-5.6 Terra.
- GPT-5.6 Terra accepts more context: 1.05M tokens versus 66K.
- WizardLM-2 8x22B has downloadable open weights; the other is API-only.
Side by side
| GPT-5.6 Terra | WizardLM-2 8x22B | |
|---|---|---|
| Provider | OpenAI | Microsoft |
| Noometry Index | 59.2 | 31.4 |
| Released | 2026-07-09 | 2024-04-15 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 66K |
| Max output | 128K | 8K |
| Input $ / M tokens | $2 | $0.62 |
| Output $ / M tokens | $12 | $0.62 |
| Results tracked | 52 | 2 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
GPT-5.6 Terra: 57.7 (#19), WizardLM-2 8x22B: —
| Benchmark | GPT-5.6 Terra | WizardLM-2 8x22B |
|---|---|---|
| DeepSWE | 69.6% | — |
| FrontierCode | 41.3% | — |
| CursorBench | 41.3% | — |
| LMArena WebDev | 1522 | — |
| SciCode | 55% | — |
| WeirdML | 78.3% | — |
| LMArena Coding | 1484 | — |
| ALE-Bench | 1,951 | — |
Agentic & Tool Use Not comparable
GPT-5.6 Terra: 40.1 (#25), WizardLM-2 8x22B: —
| Benchmark | GPT-5.6 Terra | WizardLM-2 8x22B |
|---|---|---|
| APEX-Agents | 58.2% | — |
| BALROG | 53.2% | — |
| GDP.pdf | 24.7% | — |
| Vending-Bench 2 | 7,343 | — |
Reasoning Not comparable
GPT-5.6 Terra: 60.7 (#21), WizardLM-2 8x22B: —
| Benchmark | GPT-5.6 Terra | WizardLM-2 8x22B |
|---|---|---|
| ARC-AGI-2 | 83.9% | — |
| SimpleBench | 48.9% | — |
| Kagi LLM Benchmark | 51.3% | — |
| NYT Connections (extended) | 78.4% | — |
| ARC-AGI-1 | 96.5% | — |
| CritPt | 30% | — |
| Chess Puzzles | 54% | — |
| LMArena Hard Prompts | 1468 | — |
| Mystery Game Puzzles | 35% | — |
| DTBench | 93.3% | — |
| LMCA | 55% | — |
| Surface Evolver Bench | 83.8% | — |
| Epoch Capabilities Index | 159.62 | — |
Math GPT-5.6 Terra leads
GPT-5.6 Terra: 81.6 (#12), WizardLM-2 8x22B: 28.9
| Benchmark | GPT-5.6 Terra | WizardLM-2 8x22B |
|---|---|---|
| FrontierMath (Tiers 1-3) | 86% | — |
| FrontierMath Tier 4 | 70.7% | — |
| OTIS Mock AIME 2024-2025 | 99.7% | — |
| ProofBench | 74% | — |
| LMArena Math | 1466 | — |
| MATH Level 5 | — | 25.7% |
Knowledge GPT-5.6 Terra leads
GPT-5.6 Terra: 61.2 (#30), WizardLM-2 8x22B: 23.3
| Benchmark | GPT-5.6 Terra | WizardLM-2 8x22B |
|---|---|---|
| GPQA Diamond | 93.3% | 43.4% |
| SimpleQA Verified | 43.2% | — |
| LMArena Expert | 1492 | — |
Multimodal Not comparable
GPT-5.6 Terra: 47.3 (#11), WizardLM-2 8x22B: —
| Benchmark | GPT-5.6 Terra | WizardLM-2 8x22B |
|---|---|---|
| LMArena Vision | 1271 | — |
| Blueprint-Bench 2 | 30.8% | — |
| Furniture Assembly | 54.2% | — |
| LMArena Document | 1472 | — |
Multilingual Not comparable
GPT-5.6 Terra: 54.4 (#44), WizardLM-2 8x22B: —
| Benchmark | GPT-5.6 Terra | WizardLM-2 8x22B |
|---|---|---|
| LMArena Non-English | 1439 | — |
| LMArena Chinese | 1513 | — |
| LMArena French | 1471 | — |
| LMArena German | 1460 | — |
| LMArena Japanese | 1457 | — |
| LMArena Korean | 1425 | — |
| LMArena Russian | 1450 | — |
| LMArena Spanish | 1448 | — |
Instruction Following Not comparable
GPT-5.6 Terra: 76.4 (#40), WizardLM-2 8x22B: —
| Benchmark | GPT-5.6 Terra | WizardLM-2 8x22B |
|---|---|---|
| LMArena Instruction Following | 1454 | — |
Long Context Not comparable
GPT-5.6 Terra: 44.4 (#68), WizardLM-2 8x22B: —
| Benchmark | GPT-5.6 Terra | WizardLM-2 8x22B |
|---|---|---|
| LMArena Longer Query | 1451 | — |
Writing & Preference Not comparable
GPT-5.6 Terra: 70.2 (#23), WizardLM-2 8x22B: —
| Benchmark | GPT-5.6 Terra | WizardLM-2 8x22B |
|---|---|---|
| LMArena Text | 1447 | — |
| LMArena Creative Writing | 1410 | — |
| EQ-Bench Creative Writing | 1855 | — |
| EQ-Bench 4 | 1234 | — |
| LMArena Multi-Turn | 1449 | — |
Frequently asked questions
Is GPT-5.6 Terra better than WizardLM-2 8x22B?
GPT-5.6 Terra has enough public results to be ranked (#17); WizardLM-2 8x22B does not yet, so treat this comparison as directional.
Which is cheaper, GPT-5.6 Terra or WizardLM-2 8x22B?
WizardLM-2 8x22B is cheaper. It lists at $0.62 per million input tokens and $0.62 per million output tokens; GPT-5.6 Terra lists at $2 and $12.
Which has the bigger context window?
GPT-5.6 Terra does, with 1.05M tokens against 66K.
How many benchmarks do GPT-5.6 Terra and WizardLM-2 8x22B share?
1 benchmark has published results for both models. GPT-5.6 Terra has 52 scored results on Noometry and WizardLM-2 8x22B has 2.